Inputting 1khz square at 500mV it comes out clean until I turn the volume up.
When at full volume there is pronounced ringing.
Tried modifying input filter without results.
A zobel 10r wirewound 100n on the output does little
A 10nF cap directly on the output removes the ringing.
A 3.3nF cap on the output also removes the ringing.
A 3.3nF cap with a 10ohm non inductive resistor also remove the ringing.
When at full volume there is pronounced ringing.
Tried modifying input filter without results.
A zobel 10r wirewound 100n on the output does little
A 10nF cap directly on the output removes the ringing.
A 3.3nF cap on the output also removes the ringing.
A 3.3nF cap with a 10ohm non inductive resistor also remove the ringing.
This experiments where done with a 27kohm load
With a 27ohm load the ringing is inexistent.
Now listening with AKG 32ohm headphones and producing beautiful melodies.
With a 27ohm load the ringing is inexistent.
Now listening with AKG 32ohm headphones and producing beautiful melodies.
Last edited:
The typical 10r 100n zobel distorts squares higher than 200khz, so I will settle with a 10r (non inductive) 10n zobel.
You should not be putting a 200kHz square wave through any audio equipment unless you know it is designed to handle those frequencies.
200kHz sq has frequency components that go out beyond 2MHz !
10kHz sq has frequencies that pass through normal output Zobels.
By the time you have raised this to 20kHz the Zobel will start to run warm. The resistor is starting to dissipate significant power and the ESR inside the capacitor is starting to dissipate significant power.
I would never use 50kHz square waves through an amplifier, yet many on this Forum do that and worse.
The first harmonic of 50kHz sq is 150kHz
The capacitor of an output Zobel has quite a low reactance at this frequency. Even less to the 250kHz and 350kHz of the 2nd and 3rd harmonics of the sq.
200kHz sq has frequency components that go out beyond 2MHz !
10kHz sq has frequencies that pass through normal output Zobels.
By the time you have raised this to 20kHz the Zobel will start to run warm. The resistor is starting to dissipate significant power and the ESR inside the capacitor is starting to dissipate significant power.
I would never use 50kHz square waves through an amplifier, yet many on this Forum do that and worse.
The first harmonic of 50kHz sq is 150kHz
The capacitor of an output Zobel has quite a low reactance at this frequency. Even less to the 250kHz and 350kHz of the 2nd and 3rd harmonics of the sq.
Last edited:
you cannot check for clipping/overloading when passing a sq.
You should be putting in a low level lowish frequency sine and checking you are well away from clipping and then switch to a 1kHz sq, at the same peak to peak level. That ensures the audio stages are not overloaded.
Is your gain set to 3times (+10dB)?
that makes your output ~1.5Vac ~2.12Vpk
That is a mighty high signal. Then you say you turned up to "full volume" !
Last edited:
If the simulation is accurate then that terrible peak in response @ ~10MHz is an indication something is not set up correctly. Look at how the phase plummets at that frequency. It is trying to behave like an oscillator.
Zobels now soldered in the output.
Clean squares, no ringing with 27kohm load.
Now time for listening test.
Already ordered the new opamps..... eager to listen to those also.
Clean squares, no ringing with 27kohm load.
Now time for listening test.
Already ordered the new opamps..... eager to listen to those also.
The peak also exists even with the original 33p cap in the feedback loop.
The zobel smooths things nicely (in simulation)
In post #20 you can see 20kHZ and 50kHz squarewaves when using the standard design values and a 20K pot set at half value
Nothing is peaking in my closed loop simulation of the standard circuit (33pF C2, gain 3X, and no Zobel)
Attachments
Well, I am sure it is because I did not use the most correct models and I am not an expert in simulations.
Anyway, I am now listening to your preamp and it sounds really outstanding.
I listened a bit with my entry level headphones (32ohm AKJ) and sound is really good.
I fed the preamp from the soundboard of my pc and found that it does not play very loud but did not verify the output value of the pc.... must return there latter with high spec headphones.
Anyway, I am now listening to your preamp and it sounds really outstanding.
I listened a bit with my entry level headphones (32ohm AKJ) and sound is really good.
I fed the preamp from the soundboard of my pc and found that it does not play very loud but did not verify the output value of the pc.... must return there latter with high spec headphones.
It wasn't peaking on the scope too hence the post#20 very high freq smooth sqwaves
With 3x gain, around 100dB/V headphones will do fine for max and for nice dial control
With 3x gain, around 100dB/V headphones will do fine for max and for nice dial control
For not breaking the bank I recommend you Sennheiser HD 598 or 598Se 50 Ohm (about EUR 150 best price)
Also Beyer DT 880 PRO 250 Ohm (about EUR 200 best price)
Thank you very much Salas.... I guess I will search for the Senheiser HD 598.
Do you recommend 3x gain for those 50ohm transducers ?
I have the output stage running at 100mA... should I modify that ?
Do you recommend 3x gain for those 50ohm transducers ?
I have the output stage running at 100mA... should I modify that ?